期刊
2022 IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS WORKSHOPS (ICC WORKSHOPS)
卷 -, 期 -, 页码 415-420出版社
IEEE
DOI: 10.1109/ICCWORKSHOPS53468.2022.9814634
关键词
Network intelligence; meta-learning; resource; management.
资金
- National Natural Science Foundation of China [61971061]
- Beijing Natural Science Foundation [L182039]
In this paper, the design of few-shot learning in wireless networks was studied, proposing a meta-learning model-based scheme and a coalition formation-based model selection scheme. The simulation results show that the proposed scheme can improve model accuracy performance with low communication costs.
Restricted by the data sensing capability, it is challenging for a single user to generate high-quality deep learning models based on its collected few-shot data samples. Metalearning provides a promising paradigm to make full use of historical data at the base stations to improve the performance of few-shot learning tasks. However, it is a dilemma to balance the performance and the communication costs of meta-learning. In this paper, we studied the design of few-shot learning in wireless networks. First, a meta-learning model-based scheme is designed to adapt the few-shot learning tasks, and a multicastingbased model transmission scheme is proposed. Second, a coalition formation-based model selection scheme is designed to achieve a sophisticated tradeoff between the performance and the communication costs of meta-learning. Finally, the simulation results are provided, which show that our proposed scheme can improve the model accuracy performance with low communication costs.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据